Now Assist Guardian

[nandurb]

Hello All, 

How can I locate the error message from Now Assist Guardian that blocks offensive content?

I've impersonated as an agent and started a chat via VA.

What offensive words am I typing that the agent is getting. 

Regards,

Nandu

[Dr Atul G- LNG]

Hi @nandurb 

What error you are getting?

https://www.servicenow.com/docs/bundle/yokohama-intelligent-experiences/page/administer/now-assist-p...

[nandurb]

Not getting any error. Logs are getting captured in the AI metrics table. But not blocking. How do I ensure offensive content is blocked? if blocked, how agents are getting able to see? Why below chats are not blocked?

[Dr Atul G- LNG]

https://www.servicenow.com/docs/bundle/yokohama-intelligent-experiences/page/administer/now-assist-a...

[rpriyadarshy]

Hi Nand

Here is how it works.

(1) Now Assist Guardian logs when offensive content is detected or generated when offensiveness protection is activated. You can also choose whether you want to block the content from the user. If you choose to block the content, the user sees a standardized message explaining that offensive material has been blocked instead of what was generated.

(2) Now Assist Guardian creates logs for offensive content and prompt injection guardrails. Reviewing the logs can help you determine how often offensive content is generated or prompt injection attack attempts occur.

You Can view logs for guardrails in the sys_generative_ai_metric

Procedure

1. Navigate to All > Now Assist Admin > Settings.
2. Use the more options menu item (three vertical dots) and select Export.
   1. For the offensiveness guardrail, go to Now Assist Guardian > Offensiveness.

      Use the list in the Active tab to choose which workflow you want to export logs for.

      Nice details are in below URL- 

      https://www.servicenow.com/community/now-assist-articles/now-assist-guardian-faq/ta-p/3065004

      Regards

      RP